5 papers
Taming Outlier Tokens in Diffusion Transformers
Xiaoyu Wu, Yifei Wang, Tsu-Jui Fu +3
We study outlier tokens in Diffusion Transformers (DiTs) for image generation. Prior work has shown that Vision Transformers (ViTs) can produce a small number of high-norm tokens t…
Learning Rate Engineering: From Coarse Single Parameter to Layered Evolution
Ming-Hong Yao, Di Wang, Jian Cui +5
Learning rate scheduling has evolved from the single global fixed rate of early SGD to sophisticated layer-wise adaptive strategies. We systematize this evolution into five generat…
SRSplat: Feed-Forward Super-Resolution Gaussian Splatting from Sparse Multi-View Images
Xinyuan Hu, Changyue Shi, Chuxiao Yang +6
Feed-forward 3D reconstruction from sparse, low-resolution (LR) images is a crucial capability for real-world applications, such as autonomous driving and embodied AI. However, exi…
Vision-Language-Vision Auto-Encoder: Scalable Knowledge Distillation from Diffusion Models
Tiezheng Zhang, Yitong Li, Yu-cheng Chou +4
Building state-of-the-art Vision-Language Models (VLMs) with strong captioning capabilities typically necessitates training on billions of high-quality image-text pairs, requiring…
Perception Encoder: The best visual embeddings are not at the output of the network
Daniel Bolya, Po-Yao Huang, Peize Sun +15
We introduce Perception Encoder (PE), a state-of-the-art vision encoder for image and video understanding trained via simple vision-language learning. Traditionally, vision encoder…